[PATCH v2] rust: dma: return zero for Coherent reads past EOF

[PATCH v2] rust: dma: return zero for Coherent reads past EOF

[Younes Akhouayri via B4 Relay]

From: Younes Akhouayri <git@younes.io>

Coherent<T>::write_to_slice() calculates a zero-byte copy when the file
offset is beyond the allocation, but still calls
UserSliceWriter::write_dma(). The latter rejects offsets beyond the
allocation even when the copy length is zero, so a debugfs read past EOF
returns -ERANGE.

Return before calling write_dma() when the offset is at or beyond the
allocation, matching simple_read_from_buffer() EOF semantics.

Fixes: 016818513936 ("rust: dma: implement BinaryWriter for Coherent<[u8]>")
Cc: stable@vger.kernel.org
Link: https://rust-for-linux.zulipchat.com/#narrow/channel/291566-Library/topic/.E2.9C.94.20Possible.20past-EOF.20bug.20in.20Coherent.3CT.3E.3A.3Awrite_to_slice/near/611677095
Signed-off-by: Younes Akhouayri <git@younes.io>
---
Changes in v2:
- Use ordinary subtraction after the explicit EOF check, which makes
  underflow impossible.
- Link to v1: https://patch.msgid.link/20260720-fix-dma-coherent-eof-v1-1-d92685db8da6@younes.io
---
 rust/kernel/dma.rs | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/rust/kernel/dma.rs b/rust/kernel/dma.rs
index 200def84fb69..556424b88fb9 100644
--- a/rust/kernel/dma.rs
+++ b/rust/kernel/dma.rs
@@ -1005,7 +1005,11 @@ fn write_to_slice(
             return Ok(0);
         };
 
-        let count = self.size().saturating_sub(offset_val).min(writer.len());
+        if offset_val >= self.size() {
+            return Ok(0);
+        }
+
+        let count = (self.size() - offset_val).min(writer.len());
 
         writer.write_dma(self, offset_val, count)?;
 

---
base-commit: 667d0fb32149f023b8b34a1f6f3d384556eafb5a
change-id: 20260720-fix-dma-coherent-eof-d0d08c91e013

Best regards,
--  
Younes Akhouayri <git@younes.io>

Re: [PATCH v2] rust: dma: return zero for Coherent reads past EOF

[Alexandre Courbot]

On Fri Jul 31, 2026 at 1:34 AM JST, Younes Akhouayri via B4 Relay wrote:
> From: Younes Akhouayri <git@younes.io>
>
> Coherent<T>::write_to_slice() calculates a zero-byte copy when the file
> offset is beyond the allocation, but still calls
> UserSliceWriter::write_dma(). The latter rejects offsets beyond the
> allocation even when the copy length is zero, so a debugfs read past EOF
> returns -ERANGE.
>
> Return before calling write_dma() when the offset is at or beyond the
> allocation, matching simple_read_from_buffer() EOF semantics.
>
> Fixes: 016818513936 ("rust: dma: implement BinaryWriter for Coherent<[u8]>")
> Cc: stable@vger.kernel.org
> Link: https://rust-for-linux.zulipchat.com/#narrow/channel/291566-Library/topic/.E2.9C.94.20Possible.20past-EOF.20bug.20in.20Coherent.3CT.3E.3A.3Awrite_to_slice/near/611677095
> Signed-off-by: Younes Akhouayri <git@younes.io>

Reviewed-by: Alexandre Courbot <acourbot@nvidia.com>

Re: [PATCH v2] rust: dma: return zero for Coherent reads past EOF

[Onur Özkan]

On Thu, 30 Jul 2026 18:34:38 +0200
Younes Akhouayri via B4 Relay <devnull+git.younes.io@kernel.org> wrote:

> From: Younes Akhouayri <git@younes.io>
> 
> Coherent<T>::write_to_slice() calculates a zero-byte copy when the file
> offset is beyond the allocation, but still calls
> UserSliceWriter::write_dma(). The latter rejects offsets beyond the
> allocation even when the copy length is zero, so a debugfs read past EOF
> returns -ERANGE.
> 
> Return before calling write_dma() when the offset is at or beyond the
> allocation, matching simple_read_from_buffer() EOF semantics.
> 
> Fixes: 016818513936 ("rust: dma: implement BinaryWriter for Coherent<[u8]>")
> Cc: stable@vger.kernel.org
> Link: https://rust-for-linux.zulipchat.com/#narrow/channel/291566-Library/topic/.E2.9C.94.20Possible.20past-EOF.20bug.20in.20Coherent.3CT.3E.3A.3Awrite_to_slice/near/611677095
> Signed-off-by: Younes Akhouayri <git@younes.io>
> ---
> Changes in v2:
> - Use ordinary subtraction after the explicit EOF check, which makes
>   underflow impossible.
> - Link to v1: https://patch.msgid.link/20260720-fix-dma-coherent-eof-v1-1-d92685db8da6@younes.io
> ---
>  rust/kernel/dma.rs | 6 +++++-
>  1 file changed, 5 insertions(+), 1 deletion(-)
> 
> diff --git a/rust/kernel/dma.rs b/rust/kernel/dma.rs
> index 200def84fb69..556424b88fb9 100644
> --- a/rust/kernel/dma.rs
> +++ b/rust/kernel/dma.rs
> @@ -1005,7 +1005,11 @@ fn write_to_slice(
>              return Ok(0);
>          };
>  
> -        let count = self.size().saturating_sub(offset_val).min(writer.len());

Nit: I would probably just check if `count` is zero next to this line.

> +        if offset_val >= self.size() {
> +            return Ok(0);
> +        }
> +
> +        let count = (self.size() - offset_val).min(writer.len());
>  
>          writer.write_dma(self, offset_val, count)?;
>  
> 
> ---
> base-commit: 667d0fb32149f023b8b34a1f6f3d384556eafb5a
> change-id: 20260720-fix-dma-coherent-eof-d0d08c91e013
> 
> Best regards,
> --  
> Younes Akhouayri <git@younes.io>
> 
> 

Re: [PATCH v2] rust: dma: return zero for Coherent reads past EOF

[Onur Özkan]

On Sun, 02 Aug 2026 16:54:40 +0300
Onur Özkan <work@onurozkan.dev> wrote:

> On Thu, 30 Jul 2026 18:34:38 +0200
> Younes Akhouayri via B4 Relay <devnull+git.younes.io@kernel.org> wrote:
> 
> > From: Younes Akhouayri <git@younes.io>
> > 
> > Coherent<T>::write_to_slice() calculates a zero-byte copy when the file
> > offset is beyond the allocation, but still calls
> > UserSliceWriter::write_dma(). The latter rejects offsets beyond the
> > allocation even when the copy length is zero, so a debugfs read past EOF
> > returns -ERANGE.
> > 
> > Return before calling write_dma() when the offset is at or beyond the
> > allocation, matching simple_read_from_buffer() EOF semantics.
> > 
> > Fixes: 016818513936 ("rust: dma: implement BinaryWriter for Coherent<[u8]>")
> > Cc: stable@vger.kernel.org
> > Link: https://rust-for-linux.zulipchat.com/#narrow/channel/291566-Library/topic/.E2.9C.94.20Possible.20past-EOF.20bug.20in.20Coherent.3CT.3E.3A.3Awrite_to_slice/near/611677095
> > Signed-off-by: Younes Akhouayri <git@younes.io>
> > ---
> > Changes in v2:
> > - Use ordinary subtraction after the explicit EOF check, which makes
> >   underflow impossible.
> > - Link to v1: https://patch.msgid.link/20260720-fix-dma-coherent-eof-v1-1-d92685db8da6@younes.io
> > ---
> >  rust/kernel/dma.rs | 6 +++++-
> >  1 file changed, 5 insertions(+), 1 deletion(-)
> > 
> > diff --git a/rust/kernel/dma.rs b/rust/kernel/dma.rs
> > index 200def84fb69..556424b88fb9 100644
> > --- a/rust/kernel/dma.rs
> > +++ b/rust/kernel/dma.rs
> > @@ -1005,7 +1005,11 @@ fn write_to_slice(
> >              return Ok(0);
> >          };
> >  
> > -        let count = self.size().saturating_sub(offset_val).min(writer.len());
> 
> Nit: I would probably just check if `count` is zero next to this line.

I just noticed there is already been quite a bit of discussion about this topic
so you can ignore my comment since it was just a "nit" anyway. I don't think
there's much difference either way, so there's probably no need to reignite the
discussion.

Thanks for the fix.

Reviewed-by: Onur Özkan <work@onurozkan.dev>

> 
> > +        if offset_val >= self.size() {
> > +            return Ok(0);
> > +        }
> > +
> > +        let count = (self.size() - offset_val).min(writer.len());
> >  
> >          writer.write_dma(self, offset_val, count)?;
> >  
> > 
> > ---
> > base-commit: 667d0fb32149f023b8b34a1f6f3d384556eafb5a
> > change-id: 20260720-fix-dma-coherent-eof-d0d08c91e013
> > 
> > Best regards,
> > --  
> > Younes Akhouayri <git@younes.io>
> > 
> > 

Re: [PATCH v2] rust: dma: return zero for Coherent reads past EOF

[Danilo Krummrich]

On Thu, 30 Jul 2026 18:34:38 +0200, Younes Akhouayri wrote:
> [PATCH v2] rust: dma: return zero for Coherent reads past EOF

Applied, thanks!

  Branch: driver-core-testing
  Tree:   git://git.kernel.org/pub/scm/linux/kernel/git/driver-core/driver-core.git

[1/1] rust: dma: return zero for Coherent reads past EOF
      commit: 5391b147d9fd

The patch will appear in the next linux-next integration (typically within 24
hours on weekdays).

The patch is in the driver-core-testing branch and will be promoted to
driver-core-next after validation.